Local Media Diversity Matters to All Americans

Congressional Briefing

Sponsored by the Center for American Progress

January 30^th, 2007

Remarks of Philip M. Napoli, Ph.D.

Associate Professor, Graduate School of Business

Director, Donald McGannon Communication Research Center

Fordham University

New York, NY

Thank you for the opportunity to speak here today.  My name is Philip Napoli. I am an Associate Professor of Communications and Media Management in the Graduate School of Business and the Director of the Donald McGannon Communication Research Center at Fordham University in New York.  I have had the privilege of working with Mark Lloyd and the Center for American Progress on this project.

I am going to focus on some of the logistical issues surrounding conducting the type of analyses that Mark Lloyd has presented.  Our experiences in developing a research project with this level of detail taught us some very valuable lessons about the current information environment in the area of media policy.

If we are in agreement that making effective media policy requires these kinds of detailed assessments of media markets, then there are a couple of practical realities that we need to acknowledge, and some problems that need to be corrected, for efforts such as these to successfully move forward.

The first is that, over the past 30 years, the federal government’s role in the gathering of the data necessary for conducting these kinds of analyses has diminished dramatically.

Some examples:

n      The FCC no longer gathers financial data from broadcast licensees,

n      Nor does it any longer gather any meaningful data on program practices such as the provision of local news and public affairs.

n      The gathering of data on minority and female ownership of media outlets has become increasingly haphazard and has been shown in recent studies to be incredibly flawed.

n       The FCC’s gathering of employment data has been gutted, such that it is very difficult today to get any sense of how employment in these industries is changing over time, particularly in relation to the employment of women and minorities.

All of these are categories of data that are vital to rigorously assessing issues of competition, diversity, and localism that are central to making well-informed media ownership policy decisions.

Developments such as these have coincided with a parallel development: The privatization of many of the data gathering activities that used to be conducted by the federal government.

As researchers, we are at least fortunate that there is a sizable market within the advertising, financial, and media industries for many types of data related to media industry performance.

This has meant that a number of private companies have stepped in to pick up the slack left by the FCC.

And it is these companies’ products upon which we must increasingly rely in our research, as, unfortunately, most researchers do not have nearly the resources necessary to conduct the kind of large-scale data gathering that is becoming increasingly important to producing the kind of research that has influence within the FCC, Congress, and the courts.

Unfortunately, when the gathering of such data is a function of marketplace demand, data that have commercial value are gathered quite well, while data that might have little commercial value, but tremendous policy value, are not gathered as well, if at all.

As a result, we encounter a variety of gaps in available data that make it difficult to produce studies that effectively address specific policy issues.

Thus, for example, we have found in our research that:

n      Even the largest, most comprehensive commercial databases on media ownership, media markets, and media revenues fail to gather or report data on many of the minority-targeted or foreign language newspapers in the U.S.

n      Similarly, commercial ratings often will only provide minority audience estimates for those markets with large minority populations.

n      And ratings data for many radio stations in remote, rural locations aren’t gathered or reported at all.

And so we often are operating under a very incomplete picture of the media markets we are trying to analyze because the needs of the primary clients for these commercial data providers do not correspond with the needs of policymakers or policy researchers.

However, as flawed as these data sources are, they generally are the best information that is available, and so they are still relied upon quite heavily.

Therefore, much of the research that influences the FCC’s decision-making still draws quite heavily from these sources.  The FCC itself relies quite heavily on these sources in its own work.  And so, access to these data remains critical for researchers.

Unfortunately, sufficient access to these commercial data sources is often very difficult to obtain.

One reason is that these data sources often are enormously expensive, as they are priced with commercial clients in mind as opposed to the not-for-profit policy researcher.

This means, that when it comes time to submit studies to the FCC, it is really only the large, very well funded organizations that are able to gain access to all of the best-available data.

Another reason is that these are proprietary data bases that often have very restrictive access provisions – sometimes so restrictive as to prevent collaboration or divisions of labor between researchers, or to hinder the wide dissemination of research results. 

This problem becomes compounded when the data provider is faced with the prospect of providing data to a researcher whose results could ultimately contradict the policy positions of some of the data providers’ most important clients.

Let me illustrate some of these problems with an example from research I have conducted with my colleague, Michael Yan of the University of Michigan, on local television news and public affairs programming.

The FCC traditionally has placed a tremendous emphasis on broadcasters’ provision of local television news and public fairs as a key way of both serving the needs and interests of local communities and of providing a diverse array of ideas and viewpoints to the citizenry.

Consequently, we have looked into the overall levels of local news and public affairs programming available.

We have also looked into whether there is a significant relationship between various ownership characteristics and the provision of local news and public affairs programming.

Our research has shown that more than 50% of commercial television stations in the U.S. air no local public affairs programming.

Our research has also shown that a quarter of all commercial television stations in the U.S. air no local news.

Our research also has shown that neither stations that are part of larger ownership groups, nor stations that are part of duopolies, generally convert the economies of scale that such ownership arrangements presumably allow into providing more local news and public affairs programming than other stations.

Results such as these raise some questions about broadcasters’ commitments to serving the informational needs of their local communities, as well as questions about whether we can expect any significant benefits from allowing greater concentration of ownership within individual media markets.

But what is perhaps more important is that, if we wanted to look beyond how much local news and public affairs programming is available, and consider also issues of how much of this programming actually deals with local issues, how much political news or information is provided, or even what kind of diversity of viewpoints we’re seeing on these programs, the data necessary to do this in a rigorous way are ridiculously difficult to obtain.  And to do a similar analysis for radio would be even more difficult.

For our work, we have had to rely on commercially-produced program schedule databases that are far too expensive to allow us to analyze more than a relatively small (though representative) sample of television stations, and that provide very little of the detailed content information that would allow us to look beyond basic quantities of available programming.

This is in large part because we have no comprehensive archive of local broadcast news programming in this country. 

There are instead about 40 small, poorly funded local television archives scattered across the U.S. as part of state historical societies, universities, or other entities, that generally seek to record and store limited amounts of local news programming from very confined geographic areas – and certainly aren’t conducting their work in a particularly systematic way, or with an eye towards supporting rigorous policy-related research.

There also are no federal archiving requirements for broadcast licensees, which means that broadcasters are under no obligation to make recordings of their programming available for research or public policymaking purposes.

And, getting local broadcasters to willfully provide access to local news broadcasts is very difficult, as researchers generally receive very little cooperation from broadcasters when they issue such requests.

According to a report by the Library of Congress, “Every group that has studied the selection of television for preservation has concluded that all news programs should be retained and preserved as aired,” yet according to this same report, most local newscasts are destroyed within a week.

In 1976 we saw the establishment of the American Television and Radio Archives within the Library of Congress, yet this collection remains haphazard at best.

The paradox of all this is that the FCC and the courts all continue to demand very detailed answers to very complex questions about the relationship between ownership, market conditions, and content – whether it be news, public affairs, violence, or indecency – when the information environment is completely inhospitable to doing this well.

Consider, for instance, that the FCC recently commissioned a study intended to:

“analyze the effect of ownership structure and robustness . . . on various measures of the quantity and the quality of different types of TV programming, including local news and public affairs, minority programming, children’s programming, family programming, religious programming, and violent and indecent content.”

This is an ambitious and potentially very valuable study that could provide very useful information to help in assessing current ownership policies.  Unfortunately, I can say with confidence that obtaining truly systematic and representative samples of the various kinds of programming necessary to conduct such a wide-ranging analysis is not something that can be accomplished in the contemporary information environment, in which much of the relevant data are either not gathered or are tremendously difficult to access.

This study will, instead, inevitably employ a variety of necessary shortcuts and proxies, all of which will make the results vulnerable to the intensive scrutiny of whichever stakeholder groups find themselves dissatisfied with the results.

We saw the same pattern in 2003, when very few of the studies that the FCC commissioned in connection with its 2003 ownership proceeding were able to hold up very well under intensive scrutiny.

In this way, we run the risk of our efforts to improve the state of knowledge on these issues becoming exercises in futility unless the availability and accessibility of the necessary research inputs improves dramatically.

I hope that this presentation will encourage some inquiry into the quality, quantity, and accessibility of the information that guides media policymaking.

Considering the kinds of evidentiary demands being put forth by the FCC and the courts these days, it is vital that more be done to provide researchers with access to the raw materials necessary to meet these demands.

We need the federal government to return to a more active role in the gathering of policy-relevant data, particularly in the areas of ownership, employment, revenues, and content.  Ceding all of these activities to the commercial sector means that important policy-relevant information will not be gathered effectively, if at all.

We need a system of archiving television and radio programming (particularly local news and public affairs) that is as thorough and rigorous, and as accessible, as our system of archiving newspapers.

We need a return to more rigorous reporting requirements for employment data, financial data, and programming practices.

We need policies in place that provide more equitable access to the large commercially-generated databases that increasingly serve as the backbone for research related to media ownership, diversity, and localism, in those instances when the research is being conducted for non-profit, public policy purposes.

As important as the systematic assessment of diversity and localism in media markets is to contemporary media policymaking, it is equally important that the resources available to policymakers and policy researchers to conduct such assessments are of sufficient quality, integrity and accessibility.